class AdminTagger

• AdminTagger
• Tagger
• Reference
• Object

Overview

Flags administrative / privileged endpoints. These are high-value targets for broken access control, privilege escalation, and forced browsing — surfacing them tells a reviewer where the blast radius of a missing authorization check is largest.

Defined in:

Constant Summary

READ_ONLY_METHODS = Set {"GET", "HEAD", "OPTIONS"}

STRONG_PATH_PARTS = Set {"admin", "admins", "administrator", "administration", "administrative", "superuser", "superadmin", "sysadmin", "backoffice", "impersonate", "godmode"}

Path segments that strongly imply an administrative surface. Matched as whole path segments (after splitting on /, -, _, .) so /admin/users matches but /badminton does not. /wp-admin and /super-admin are covered too: the split yields an admin token. superadmin (no separator) is listed explicitly since the split can't recover it.

STRONG_PRIVILEGE_PARAM_NAMES = Set {"is_admin", "is_superuser", "is_superadmin", "is_staff", "is_root", "make_admin", "grant_admin", "admin_only", "superadmin", "sudo", "impersonate"}

Parameter names that imply a privilege/role grant regardless of the route or method, e.g. a generic /users/{id} PATCH that accepts is_admin. These are specific enough to flag on their own. Matched separator-insensitively via normalize_param_name, so is_admin, isAdmin, and is-admin all collapse to the same key.

STRONG_PRIVILEGE_PARAM_NAMES_NORMALIZED = STRONG_PRIVILEGE_PARAM_NAMES.map(&.gsub(/[-_]/, "")).to_set

WEAK_PRIVILEGE_PARAM_NAMES = Set {"run_as", "as_user", "elevate", "privilege", "privileged"}

Weaker, more generic privilege hints. These also appear as read-only filters (GET /roles?privilege=x, ?as_user=... view switching), so only flag them on a state-changing (non-read) method.

WEAK_PRIVILEGE_PARAM_NAMES_NORMALIZED = WEAK_PRIVILEGE_PARAM_NAMES.map(&.gsub(/[-_]/, "")).to_set

Constructors

• .new(options : Hash(String, YAML::Any))

Instance Method Summary

• #perform(endpoints : Array(Endpoint))

Constructor Detail

Instance Method Detail